the recent increase of methane from 10 years of ndacc … back up... · the recent increase of...

The recent increase of methanefrom 10 years of NDACC

ground-based FTIR observationsBader, W. et al., Ten years of atmospheric methane from ground-based NDACC FTIR observations, Atmos. Chem. Phys. Discuss.,

doi:10.5194/acp-2016-699, in review, 2016.

Logo EU

Atmospheric methane• 2nd most important anthropogenic greenhouse gas

• 1/5 of anthropogenic radiative forcing since 1750 is due to methane

• 3 types of emission processes :• biogenic (dotted)

• thermogenic (plain)

• pyrogenic (hatched)

• 1 major sink

• oxidation by OH

+ 260% since 1750

Ace Science Team Meeting - 24 Oct 2016 2

Trends of methane in the last 20 years


0.54 ± 0.18 %

0.04 ± 0.08 %

0.31 ± 0.04 %

Jungfraujoch : Swiss Alps – Remote site – Northern mid-latitudeSolar observations from Fourier Transform Infrared spectrometer

Unexplained increase since ~2005


0.54 ± 0.18 %

-0.04 ± 0.08 %

0.31 ± 0.04 %

10 years of NDACC FTIR observations


1- Eureka (80 °N)2- Kiruna (68 °N)3- Zugspitze (47 °N)4- Jungfraujoch (47 °N)5- Toronto (44 °N)

6- Tsukuba (37 °N)7- Izaña (28 °N)8- Wollongong (34 °S)9- Lauder (45 °S)10- Arrival Heights (78 °S)

2005-2014

10 years of NDACC FTIR observations


> Anomaly with respect to 2005.0 Averaged increase : 0.31 ± 0.03 %/year

0.26 ± 0.02 %/year : Wollongong0.39 ± 0.09 %/year : Toronto

• Resolution : 2° x 2.5° and 4° x 5°• 47 vertical levels - Output : 3 hours• Meteo fields : GEOS5 (Dec 2003 -

May 2013)• Spin-up over 2004 (70 spins for

initialization)• Emission inventories

o Anthropogenic emissions :EDGAR v4.2

o Biomass burning : GFED3 (8h)o Wetland model [Pikett-Heaps,

2011]o Termites [Fung et al., 1991]o Biofuels [Yevich and Logan, 2003]o Soil absorption [Fung et al., 1991]

• Main sink : 3D OH monthly [Park etal., 2004]

• Lifetime : 8.9 years

Each tracer represents thecontribution of each source to thesimulated total column of methane

Tracers

1- Total

2- Gas and oil

3- Coal

4- Livestock

5- Waste management

6- Biofuels

7- Rice cultures

8- Biomass burning

9- Wetlands

10- Other natural

11- Other anthropogenic

12- Soil absorption


GEOS-Chem v9-02. Tagged simulation

FTIR vs GEOS-Chem


• Maximum bias 4.8 ± 3.5 % (Arrival Heights) ≈ FTIR systematic error

> Anomaly with respect to 2005.0

FTIR vs GEOS-Chem


Averaged increase over 2005-2012 FTIR : 0.30 ± 0.04 %/year

GEOS-Chem : 0.35 ± 0.03 %/year

Methane since 2005


GC tagged simulation : Analysis


Yearly relative change in %.

𝑌𝐶 𝑖𝑛 % =𝜇𝑛 − 𝜇𝑛−1𝜇𝑡𝑜𝑡,𝑛−1

μn : annual mean of CH4, year n.

The year-to-year relativechanges are computed so thatwhen we assume a relativechange of a tracer for the year n,it is expressed wrt to theprevious year (n-1) as reference.

µtot, n-1 : annual mean of thesimulated cumulative methanefor the year (n-1)

GC tagged simulation : Analysis• Yearly relative change (%)

• Natural sources mainly responsible for the interannual variation

• e.g. Wollongong


GC tagged simulation : Analysis• Yearly relative change (%)

• Secondary contributors to the budget contribute to the increase

• e.g. Izaña


GC tagged simulation : Analysis


EUR KIR ZUG JFJ TOR TSU IZA WOL LAU AHT

1 co co co co co co co co co co

2 gao gao wl gao gao gao gao li gao gao

3 wl wl gao wl wl wl wl gao li li

4 ri li ri ri ri li li wl wl wl

5 li ri li li li ri ri ri ri ri

6 wa wa wa wa wa wa wa wa wa wa

7 bf bf oa oa bf bf bf bf bf bf

8 oa oa bf bf oa oa oa oa oa oa

9 on on on on on on on on on on

10 bb bb bb bb bb bb bb bb bb bb

• Contributors to the simulated increase : tracer ranking

Methane increase : Discussion

Many studies…Rigby et al., 2008Ringeval et al., 2010Bloom et al., 2010

Aydin et al., 2011Dlugokencky et al., 2009Sussmann et al., 2012

Kirschke et al., 2013Nisbet et al., 2014Hausmann et al., 2016Schaefer et al., 2016


Source attribution ?


Many studies…Rigby et al., 2008Ringeval et al., 2010Bloom et al., 2010

Aydin et al., 2011Dlugokencky et al., 2009Sussmann et al., 2012

Kirschke et al., 2013Nisbet et al., 2014Hausmann et al., 2016Schaefer et al., 2016


On the sink side,

Rigby et al. (2008) OH − 4 ± 14 % from 2006 to 2007

Montzka et al. (2011) -1 %/year

Bousquet et al. (2011) < 1 % over the 2006-2008 time period

Dentener (2003)

“atmospheric composition generally buffer the global OH concentrations”



From GEOS-Chem tagged simulationSecondary contributors to the global budget of methane play amajor role in the increase of methane observed since 2005.→ coal mining, gas and oil transport and exploitation




From GEOS-Chem tagged simulationSecondary contributors to the global budget of methane play amajor role in the increase of methane observed since 2005.→ coal mining, gas and oil transport and exploitation

Best emission inventories available → limitations

EDGAR v4.2

• Overestimates the recent emission growth in Asia (Schwietzke et al. 2014, Bergamaschi et al. 2013 and Bruhwiler et al. 2014).

• Chinese coal mining emissions are too large by a factor of 2 (Turner et al. 2015, from a global GOSAT inversion)

• EDGAR v4.2 vs global GOSAT inversion (Turner et al., 2015)• increase in wetland emissions in South America

• increase in rice emissions in Southeast Asia




Gas and oil use and exploitation (GAO)

underestimated by current emission inventories (incl. EDGAR)

Franco et al., 2015, 2016; Turner et al., 2015, 2016




Gas and oil emissions : the use of C2H6 as a proxy

Production, transport and use of natural gas and the leakage associated to itamounts at ~ 62 % of ethane’s atmospheric budget (Logan et al., 1981;Rudolph, 1995)






Franco et al., 2016

• Observations : sharp increase of C2H6 since 2009

→ ~5 %/year at mid-latitudes, ~3 %/year at remote sites

• Massive growth of oil and gas exploitation in the North Americancontinent, confirmed by Helmig et al. 2016






Franco et al., 2016

• Observations : sharp increase of C2H6 since 2009

→ ~5 %/year at mid-latitudes, ~3 %/year at remote sites

• Massive growth of oil and gas exploitation in the North Americancontinent, confirmed by Helmig et al. 2016

Methane increase ? From C2H6/CH4 ratio derived from GOSATmeasurements

• → from 20 Tg in 2008, to 35 Tg in 2014

• Confirming the influence of GAO on the observed methane increase






Hausmann et al. 2016

C2H6/CH4 ratio derived from an atmospheric two-box model

GAO contribution of 39 % to the renewed methane in Zugspitze between2007 and 2014






Hausmann et al. 2016

C2H6/CH4 ratio derived from an atmospheric two-box model

GAO contribution of 39 % to the renewed methane in Zugspitze between2007 and 2014

BUT

The strength of the C2H6/CH4 relationship associated to GAO stronglydepends on the studied region and/or production basin

Variability rarely taken into account (Kort et al. 2016, Peischl et al. 2016)







Lyon et al. (2016) pointed out that emissions from GAO well pads may be missing from most bottom-up emission inventories.



A horizontal drilling rig for natural gas in the Marcellus formation in

eastern Lycoming County, Pennsylvania.





Lyon et al. (2016) pointed out that emissions from GAO well pads may be missing from most bottom-up emission inventories.



Emissions differ from one well pad toanother and even within the samepad depending on the depth of theextraction.

e.g. Marcellus Basin that is actually two different overlapping basins. The depth of the well thus determines into which basin it’s drilling.

Conclusions

• CH4 change since 2005 from FTIR ground-based measurements

• 0.31 ± 0.03 %/year with respect to 2005.0

• Consistent with methane changes computed from in situ measurements


• GEOS-Chem tagged simulation : 11 tracers from 2005 to 2012

• 0.35 ± 0.03 %/year vs 0.30 ± 0.04 %/year (FTIR)

• Natural sources contribute to the inter-annual variability

• Anthropogenic sources, second contributors to the global budgetmethane, contribute to the observed increase• coal mining

• gas and oil exploitation

• livestock


Conclusions

• While we showed that GEOS-Chem agrees with our observations, therepartition between the different sources of methane would greatlybenefit from an improvement of the global emission inventories.

• EDGAR v4.2• US oil and gas and livestock are underestimated.

• Coal emissions are overestimated.

• Further attention has to be given to improved anthropogenic methaneinventories, such as emission inventories associated with fossil fuel andnatural gas production. This is essential in a context of the energytransition that includes the development of shale gas exploitation.


THANKYOU !

Bader, W. et al., Ten years of atmospheric methane from ground-based NDACC FTIR observations, Atmos. Chem. Phys. Discuss.,

doi:10.5194/acp-2016-699, in review, 2016.

The University of Liège’s involvement has primarily been supported by the PRODEX and SSD programs funded by theBelgian Federal Science Policy Office (Belspo), Brussels. W.B. has received support by a postdoctoral fellowship fromthe University of Toronto. This project has received funding from the European Union’s Horizon 2020 research andinnovation programme under the Marie Sklodowska-Curie grant agreement No 704951.

the recent increase of methane from 10 years of ndacc … back up... · the recent increase of...

Documents